Independent-component analysis of skin color image

Integrated deep learning approach for generating cross-polarized images and analyzing skin melanin and hemoglobin distributions

Cross-polarized images are beneficial for skin pigment analysis due to the enhanced visualization of melanin and hemoglobin regions. However, the required imaging equipment can be bulky and optically complex. Additionally, preparing ground truths for training pigment analysis models is labor-intensive. This study aims to introduce an integrated approach for generating cross-polarized images and creating skin melanin and hemoglobin maps without the need for ground truth preparation for pigment distributions. We propose a two-component approach: a cross-polarized image generation module and a skin analysis module. Three generative adversarial networks (CycleGAN, pix2pix, and pix2pixHD) are compared for creating cross-polarized images. The regression analysis network for skin analysis is trained with theoretically reconstructed ground truths based on the optical properties of pigments. The methodology is evaluated using the VISIA VAESTRO clinical system. The cross-polarized image generation module achieved a peak signal-to-noise ratio of 35.514 dB. The skin analysis module demonstrated correlation coefficients of 0.942 for hemoglobin and 0.922 for melanin. The integrated approach yielded correlation coefficients of 0.923 for hemoglobin and 0.897 for melanin, respectively. The proposed approach achieved a reasonable correlation with the professional system using actually captured images, offering a promising alternative to existing professional equipment without the need for additional optical instruments or extensive ground truth preparation.

Blind Separation of Skin Chromophores from Multispectral Dermatological Images

Background/Objectives: Based on Blind Source Separation and the use of multispectral imaging, the new approach we propose in this paper aims to improve the estimation of the concentrations of the main skin chromophores (melanin, oxyhemoglobin and deoxyhemoglobin), while considering shading as a fully-fledged source. Methods: In this paper, we demonstrate that the use of the Infra-Red spectral band, in addition to the traditional RGB spectral bands of dermatological images, allows us to model the image provided by each spectral band as a mixture of the concentrations of the three chromophores in addition to that of the shading, which are estimated through four steps using Blind Source Separation. Results: We studied the performance of our new method on a database of real multispectral dermatological images of melanoma by proposing a new quantitative performances measurement criterion based on mutual information. We then validated these performances on a database of multispectral dermatological images that we simulated using our own new protocol. Conclusions: All the results obtained demonstrated the effectiveness of our new approach for estimating the concentrations of the skin chromophores from a multispectral dermatological image, compared to traditional approaches that consist of using only the RGB image by neglecting shading.

Rapid low-cost hyperspectral imaging system for quantitative assessment of infantile hemangioma

Hemangioma, the predominant benign tumor occurring in infancy, exhibits a wide range of prognoses and associated outcomes. The accurate determination of prognosis through noninvasive imaging modalities holds essential importance in enabling effective personalized treatment strategies and minimizing unnecessary surgical interventions for individual patients. The present study focuses on advancing the personalized prognosis of hemangioma by leveraging noninvasive optical sensing technologies by the development of a novel rapid hyperspectral sensor (image collection in 5 s, lateral resolution of 10 μm) that is capable of quantifying hemoglobin oxygenation and vascularization dynamics during the course of tumor evolution. We have developed a quantitative parameter for hemangioma assessment, that demonstrated agreement with the clinician's conclusion in 90% among all cases during clinical studies on six patients, who visited clinician from two to four times. The presented methodology has potential to be implemented as a supportive tool for accurate hemangioma diagnostics in clinics.

Skin optical properties from 200 to 300 nm support far UV-C skin-safety in vivo

The growing threat of multi-drug resistant pathogens and airborne microbial diseases has highlighted the need to improve or develop novel disinfection methods for clinical environments. Conventional ultraviolet C (UV-C) lamps effectively inactivate microorganisms but are harmful to human skin and eyes upon exposure. The use of new 233 nm far UV-C LEDs as an antiseptic can overcome those limitations. In this research, the light penetration into the skin was elucidated for the UV-C region (<300 nm) by measuring the scattering and absorption of skin layers and inverse Monte Carlo simulation, and further confirmed by the first clinical pilot trial in which healthy volunteers were irradiated with a dose of 60 mJ/cm2 at 233 nm. The radiation is strongly absorbed in the stratum corneum, resulting in minimal skin damage without inducing inflammatory responses. The results suggest that 233 nm far UV-C light emitting diodes (LEDs) could effectively inactivate microorganisms, while being safe and soft for the skin.

Epidemiological survey of the quantity and anatomical position of facial pigmented spots in relation to age in 454 Japanese women

Facial pigmented spots are one of the phenotypes of skin aging, but no quantitative analysis of spot features such as color intensity, size, anatomical position, and number on the cheek has yet been performed. In the current study, we conducted an epidemiological survey of 454 Japanese women in their 20s to 70s and analyzed age-related changes and site differences of facial pigmented spots. Using image analysis of high-resolution digital facial photographs, 4912 individual pigmented spots were quantified according to color, size, anatomical position, and total number on the cheek. As a result of color analysis, the color intensity of individual pigmented spots increased with aging, significantly so between ages 30s and 50s. The age-related increase in melanin index of facial spots was confirmed in all sites but did not significantly differ between facial subregions. Regarding the size of pigmented spots, the frequency of large spots increased with age, and large spots were detected in all facial sites. The total number of pigmented spots on the entire cheek increased with aging, significantly so between the 20s and 40s. The number of pigmented spots tended to increase from the region near the canthi to the lower cheeks. The number of spots was markedly increased in the buccal regions compared with the infraorbital and zygomatic regions. The data and methodology presented in the current study can link the state of facial pigmentation with the various factors involved in the histological development of pigmented spots, opening new possibilities in the fields of skin pharmacology and dermatology.

An EfficientNet-based modified sigmoid transform for enhancing dermatological macro-images of melanoma and nevi skin lesions

BACKGROUND AND OBJECTIVE

During the initial stages, skin lesions may not have sufficient intensity difference or contrast from the background region on dermatological macro-images. The lack of proper light exposure at the time of capturing the image also reduces the contrast. Low contrast between lesion and background regions adversely impacts segmentation. Enhancement techniques for improving the contrast between lesion and background skin on dermatological macro-images are limited in the literature. An EfficientNet-based modified sigmoid transform for enhancing the contrast on dermatological macro-images is proposed to address this issue.

METHODS

A modified sigmoid transform is applied in the HSV color space. The crossover point in the modified sigmoid transform that divides the macro-image into lesion and background is predicted using a modified EfficientNet regressor to exclude manual intervention and subjectivity. The Modified EfficientNet regressor is constructed by replacing the classifier layer in the conventional EfficientNet with a regression layer. Transfer learning is employed to reduce the training time and size of the dataset required to train the modified EfficientNet regressor. For training the modified EfficientNet regressor, a set of value components extracted from the HSV color space representation of the macro-images in the training dataset is fed as input. The corresponding set of ideal crossover points at which the values of Dice similarity coefficient (DSC) between the ground-truth images and the segmented output images obtained from Otsu's thresholding are maximum, is defined as the target.

RESULTS

On images enhanced with the proposed framework, the DSC of segmented results obtained by Otsu's thresholding increased from 0.68 ± 0.34 to 0.81 ± 0.17.

CONCLUSIONS

The proposed algorithm could consistently improve the contrast between lesion and background on a comprehensive set of test images, justifying its applications in automated analysis of dermatological macro-images.

Fabrication of a Human Skin Mockup with a Multilayered Concentration Map of Pigment Components Using a UV Printer

In this paper, we propose a pipeline that reproduces human skin mockups using a UV printer by obtaining the spatial concentration map of pigments from an RGB image of human skin. The pigment concentration distributions were obtained by a separating method of skin pigment components with independent component analysis from the skin image. This method can extract the concentration of melanin and hemoglobin components, which are the main pigments that make up skin tone. Based on this concentration, we developed a procedure to reproduce a skin mockup with a multi-layered structure that is determined by mapping the absorbance of melanin and hemoglobin to CMYK (Cyan, Magenta, Yellow, Black) subtractive color mixing. In our proposed method, the multi-layered structure with different pigments in each layer contributes greatly to the accurate reproduction of skin tones. We use a UV printer because the printer is capable of layered fabrication by using UV-curable inks. As the result, subjective evaluation showed that the artificial skin reproduced by our method has a more skin-like appearance than that produced using conventional printing.

Visualizing veins from color images under varying illuminations for medical applications

SIGNIFICANCE

Effective vein visualization is critically important for several clinical procedures, such as venous blood sampling and intravenous injection. Existing technologies using infrared device or ultrasound rely on professional equipment and are not suitable for daily medical care. A regression-based vein visualization method is proposed.

AIM

We visualize veins from conventional RGB images to provide assistance in venipuncture procedures as well as clinical diagnosis of some venous insufficiency.

APPROACH

The RGB images taken by digital cameras are first transformed to spectral reflectance images using Wiener estimation. Multiple regression analysis is then applied to derive the relationship between spectral reflectance and the concentrations of pigments. Monte Carlo simulation is adopted to get prior information. Finally, vein patterns are visualized from the spatial distribution of pigments. To minimize the effect of illumination on skin color, light correction and shading removal operations are performed in advance.

RESULTS

Experimental results from inner forearms of 60 subjects show the effectiveness of the regression-based method. Subjective and objective evaluations demonstrate that the clarity and completeness of vein patterns can be improved by light correction and shading removal.

CONCLUSIONS

Vein patterns can be successfully visualized from RGB images without any professional equipment. The proposed method can assist in venipuncture procedures. It also shows promising potential to be used in clinical diagnosis and treatment of some venous insufficiency.

Segmenting Vitiligo on Clinical Face Images Using CNN Trained on Synthetic and Internet Images

Accurately diagnosing and describing the severity of vitiligo is crucial for prognostication, treatment selection and comparison. Currently, disease severity scores require dermatologists to estimate percentage area of involvement, which is subjected to inter and intra-assessor variability. Previous studies focus on pure skin but vitiligo on the face, which has a more serious impact on patients' quality of life, was completely neglected. Convolutional neural networks (CNNs) have good performance on many segmentation tasks. However, due to data privacy, it is hard to have a large clinical vitiligo face image dataset to train a CNN. To address this challenge, images from two different sources, the Internet and the proposed vitiligo face synthesis algorithm, are employed in training. 843 vitiligo images taken from different viewpoints were collected from the Internet. These images are hugely different from the target clinical images collected according to a newly established international standard. To have more vitiligo face images similar to the target clinical images to enhance segmentation performance, an image synthesis algorithm is proposed. Both synthetic and Internet images are used to train a CNN which is modified from the fully convolutional network (FCN) to segment face vitiligo lesions. The results show that 1) the synthetic images effectively improve segmentation performance; 2) the proposed algorithm achieves 1.06 % error for the face vitiligo area estimation and 3) it is more accurate than two dermatologists and all the previous automated vitiligo segmentation methods, which were designed for segmentation vitiligo on pure skin.

AI-based localization and classification of skin disease with erythema

Although computer-aided diagnosis (CAD) is used to improve the quality of diagnosis in various medical fields such as mammography and colonography, it is not used in dermatology, where noninvasive screening tests are performed only with the naked eye, and avoidable inaccuracies may exist. This study shows that CAD may also be a viable option in dermatology by presenting a novel method to sequentially combine accurate segmentation and classification models. Given an image of the skin, we decompose the image to normalize and extract high-level features. Using a neural network-based segmentation model to create a segmented map of the image, we then cluster sections of abnormal skin and pass this information to a classification model. We classify each cluster into different common skin diseases using another neural network model. Our segmentation model achieves better performance compared to previous studies, and also achieves a near-perfect sensitivity score in unfavorable conditions. Our classification model is more accurate than a baseline model trained without segmentation, while also being able to classify multiple diseases within a single image. This improved performance may be sufficient to use CAD in the field of dermatology.

Facial UV photo imaging for skin pigmentation assessment using conditional generative adversarial networks

Skin pigmentation is associated with skin damages and skin cancers, and ultraviolet (UV) photography is used as a minimally invasive mean for the assessment of pigmentation. Since UV photography equipment is not usually available in general practice, technologies emphasizing pigmentation in color photo images are desired for daily care. We propose a new method using conditional generative adversarial networks, named UV-photo Net, to generate synthetic UV images from color photo images. Evaluations using color and UV photo image pairs taken by a UV photography system demonstrated that pigment spots were well reproduced in synthetic UV images by UV-photo Net, and some of the reproduced pigment spots were difficult to be recognized in color photo images. In the pigment spot detection analysis, the rate of pigment spot areas in cheek regions for synthetic UV images was highly correlated with the rate for UV photo images (Pearson's correlation coefficient 0.92). We also demonstrated that UV-photo Net was effective for floating up pigment spots for photo images taken by a smartphone camera. UV-photo Net enables an easy assessment of pigmentation from color photo images and will promote self-care of skin damages and early signs of skin cancers for preventive medicine.

Automatic Classification of Adult Males With and Without Autism Spectrum Disorder by Non-contact Measurement of Autonomic Nervous System Activation

People with autism spectrum disorder (ASD) exhibit atypicality in various domains of behavior. Previous psychophysiological studies have revealed an atypical pattern of autonomic nervous system (ANS) activation induced by psychosocial stimulation. Thus, it might be feasible to develop a novel assessment tool to evaluate the risk of ASD by measuring ANS activation in response to emotional stimulation. The present study investigated whether people with ASD could be automatically classified from neurotypical adults based solely on physiological data obtained by the recently introduced non-contact measurement of pulse wave. We video-recorded faces of adult males with and without ASD while watching emotion-inducing video clips. Features reflective of ANS activation were extracted from the temporal fluctuation of facial skin coloration and entered into a machine-learning algorithm. Though the performance was modest, the gradient boosting classifier succeeded in classifying people with and without ASD, which indicates that facial skin color fluctuation contains information useful for detecting people with ASD. Taking into consideration the fact that the current study recruited only high-functioning adults who have relatively mild symptoms and probably developed some compensatory strategies, ASD screening by non-contact measurement of pulse wave could be a promising assessment tool to evaluate ASD risk.

Simple and affordable imaging of multiple physiological parameters with RGB camera-based diffuse reflectance spectroscopy

We propose a simple and affordable imaging technique to evaluate transcutaneously multiple physiological parameters by using a digital red-green-blue camera. In this method, the RGB-values were converted into tristimulus values in the CIE (Commission Internationale de l'Eclairage) XYZ color space, which is compatible with the common color spaces. Monte Carlo simulation for light transport in biological tissue was then performed to specify the relationship among the XYZ-values and the concentrations of oxygenated hemoglobin, deoxygenated hemoglobin, bilirubin, and melanin. The concentration of total hemoglobin and tissue oxygen saturation were also calculated from the estimated concentrations of oxygenated and deoxygenated hemoglobin. In vivo experiments with bile duct ligation in rats demonstrated that the estimated bilirubin concentration increased after ligation of the bile duct and reached around 22 mg/dl at 116 h after the onset of ligation, which corresponds to the ground truth value of bilirubin measured by a commercially available transcutaneous bilirubinometer. Experiments with rats while varying the fraction of inspired oxygen demonstrated that oxygenated hemoglobin and deoxygenated hemoglobin decreased and increased, respectively, as the fraction of inspired oxygen decreased. Consequently, tissue oxygen saturation dramatically decreased. We further extended the method to a non-contact imaging photo-plethysmograph and estimation of the percutaneous oxygen saturation. An empirical formula to estimate percutaneous oxygen saturation was derived from the pulse wave amplitudes of oxygenated and deoxygenated hemoglobin. The estimated percutaneous oxygen saturation dropped remarkably when a faction of inspired oxygen was below 19%, indicating the onset of hypoxemia due to hypoxia, whereas the tissue oxygen saturation decreased gradually according to the reduction of the faction of inspired oxygen. The results in this study indicate the potential of this method for imaging of multiple physiological parameters in skin tissue and evaluating an optical biomedical imaging technique that enables cost-effective, easy-to-use, portable, remotely administered, and/or point-of-care solutions.

Three-dimensional maps of human skin properties on full face with shadows using 3-D hyperspectral imaging

Hyperspectral imaging has shown great potential for optical skin analysis by providing noninvasive, pixel-by-pixel surface measurements from which, applying an optical model, information such as melanin concentration and total blood volume fraction can be mapped. Such applications have been successfully performed on small flat skin areas, but existing methods are not suited to large areas such as an organ or a face, due to the difficulty of ensuring homogeneous illumination on complex three-dimensional (3-D) objects, which leads to errors in the maps. We investigate two methods to account for these irradiance variations on a face. The first one relies on a radiometric correction of the irradiance, using 3-D information on the face's shape acquired by combining the hyperspectral camera with a 3-D scanner; the second relies on an optimization metric used in the map computation, which is invariant to irradiance. We discuss the advantages and drawbacks of the two methods, after having presented in detail the whole acquisition setup, which has been designed to provide high-resolution images with a short acquisition time, as required for live surface measurements of complex 3-D objects such as the face.

Automatic psoriasis lesion segmentation in two-dimensional skin images using multiscale superpixel clustering

Psoriasis is a chronic skin disease that is assessed visually by dermatologists. The Psoriasis Area and Severity Index (PASI) is the current gold standard used to measure lesion severity by evaluating four parameters, namely, area, erythema, scaliness, and thickness. In this context, psoriasis skin lesion segmentation is required as the basis for PASI scoring. An automatic lesion segmentation method by leveraging multiscale superpixels and [Formula: see text]-means clustering is outlined. Specifically, we apply a superpixel segmentation strategy on CIE-[Formula: see text] color space using different scales. Also, we suppress the superpixels that belong to nonskin areas. Once similar regions on different scales are obtained, the [Formula: see text]-means algorithm is used to cluster each superpixel scale separately into normal and lesion skin areas. Features from both [Formula: see text] and [Formula: see text] color bands are used in the clustering process. Furthermore, majority voting is performed to fuse the segmentation results from different scales to obtain the final output. The proposed method is extensively evaluated on a set of 457 psoriasis digital images, acquired from the Royal Melbourne Hospital, Melbourne, Australia. Experimental results have shown evidence that the method is very effective and efficient, even when applied to images containing hairy skin and diverse lesion size, shape, and severity. It has also been ascertained that CIE-[Formula: see text] outperforms other color spaces for psoriasis lesion analysis and segmentation. In addition, we use three evaluation metrics, namely, Dice coefficient, Jaccard index, and pixel accuracy where scores of 0.783%, 0.698%, and 86.99% have been achieved by the proposed method for the three metrics, respectively. Finally, compared with existing methods that employ either skin decomposition and support vector machine classifier or Euclidean distance in the hue-chrome plane, our multiscale superpixel-based method achieves markedly better performance with at least 20% accuracy enhancement.

Noncontact measurement of emotional and physiological changes in heart rate from a webcam

Heart rate, measured in beats per minute, can be used as an index of an individual's physiological state. Each time the heart beats, blood is expelled and travels through the body. This blood flow can be detected in the face using a standard webcam that is able to pick up subtle changes in color that cannot be seen by the naked eye. Due to the light absorption spectrum of blood, we are able to detect differences in the amount of light absorbed by the blood traveling just below the skin (i.e., photoplethysmography). By modulating emotional and physiological stress-that is, viewing arousing images and sitting versus standing, respectively-to elicit changes in heart rate, we explored the feasibility of using a webcam as a psychophysiological measurement of autonomic activity. We found a high level of agreement between established physiological measures, electrocardiogram, and blood pulse oximetry, and heart rate estimates obtained from the webcam. We thus suggest webcams can be used as a noninvasive and readily available method for measuring psychophysiological changes, easily integrated into existing stimulus presentation software and hardware setups.

Evaluation of non-invasive multispectral imaging as a tool for measuring the effect of systemic therapy in Kaposi sarcoma

Diffuse multi-spectral imaging has been evaluated as a potential non-invasive marker of tumor response. Multi-spectral images of Kaposi sarcoma skin lesions were taken over the course of treatment, and blood volume and oxygenation concentration maps were obtained through principal component analysis (PCA) of the data. These images were compared with clinical and pathological responses determined by conventional means. We demonstrate that cutaneous lesions have increased blood volume concentration and that changes in this parameter are a reliable indicator of treatment efficacy, differentiating responders and non-responders. Blood volume decreased by at least 20% in all lesions that responded by clinical criteria and increased in the two lesions that did not respond clinically. Responses as assessed by multi-spectral imaging also generally correlated with overall patient clinical response assessment, were often detectable earlier in the course of therapy, and are less subject to observer variability than conventional clinical assessment. Tissue oxygenation was more variable, with lesions often showing decreased oxygenation in the center surrounded by a zone of increased oxygenation. This technique could potentially be a clinically useful supplement to existing response assessment in KS, providing an early, quantitative, and non-invasive marker of treatment effect.

A coarse-to-fine approach for segmenting melanocytic skin lesions in standard camera images

Melanoma is a type of malignant melanocytic skin lesion, and it is among the most life threatening existing cancers if not treated at an early stage. Computer-aided prescreening systems for melanocytic skin lesions is a recent trend to detect malignant melanocytic skin lesions in their early stages, and lesion segmentation is an important initial processing step. A good definition of the lesion area and its border is very important for discriminating between benign and malignant cases. In this paper, we propose to segment melanocytic skin lesions using a sequence of steps. We start by pre-segmenting the skin lesion, creating a new image representation (channel) where the lesion features are more evident. This new channel is thresholded, and the lesion border pre-detection is refined using an active-contours algorithm followed by morphological operations. Our experimental results based on a publicly available dataset suggest that our method potentially can be more accurate than comparable state-of-the-art methods proposed in literature.

3-D volume reconstruction of skin lesions for melanin and blood volume estimation and lesion severity analysis

Subsurface information about skin lesions, such as the blood volume beneath the lesion, is important for the analysis of lesion severity towards early detection of skin cancer such as malignant melanoma. Depth information can be obtained from diffuse reflectance based multispectral transillumination images of the skin. An inverse volume reconstruction method is presented which uses a genetic algorithm optimization procedure with a novel population initialization routine and nudge operator based on the multispectral images to reconstruct the melanin and blood layer volume components. Forward model evaluation for fitness calculation is performed using a parallel processing voxel-based Monte Carlo simulation of light in skin. Reconstruction results for simulated lesions show excellent volume accuracy. Preliminary validation is also done using a set of 14 clinical lesions, categorized into lesion severity by an expert dermatologist. Using two features, the average blood layer thickness and the ratio of blood volume to total lesion volume, the lesions can be classified into mild and moderate/severe classes with 100% accuracy. The method therefore has excellent potential for detection and analysis of pre-malignant lesions.

Use of digital patient photographs and electronic medical record data as diagnostic tools in Japan

An electronic medical record (EMR) system was introduced to the University of Miyazaki Hospital, in Japan, in 2006. This hospital is the only one in Japan to store digital photographs of patients within EMRs. In this paper, we report on the utility of these digital photographs for disease diagnosis. Digital photographs of patients were taken at the time of hospitalization, and have been used for patient identification by medical staff. More than 20,000 digital photographs have been saved, along with examination data and medical history classified by disease, since the introduction of EMR. In the first part of the present study, we analyzed the facial cheek color of patients using photographs taken at the time of hospitalization in relation to diagnoses in six disease categories that were considered to lead to characteristic facial skin characteristics. We verified the presence or absence of a characteristic color for each disease category. Next, we focused on four diseases, Analysis of the facial skin color of 1268 patients found the same patterns of characteristic color. Overall, we found significant differences in complexion according to disease type, based on the analysis of color from digital photos and other EMR information. We propose that color analysis data should become an additional item of information stored in EMRs.

Computerised image analysis of vitiligo lesion: evaluation using manually defined lesion areas

BACKGROUND

Vitiligo is a cutaneous pigmentary disorder characterized by depigmented macules and patches that result from loss of epidermal melanocytes. Physician evaluates the efficacy of treatment by comparing the extent of vitiligo lesions before and after treatment based on the overall visual impression of the treatment response. This method is called the physician's global assessment (PGA) which is subjective. In this article, we present an innovative digital image processing method to determine vitiligo lesion area in an objective manner.

METHOD

The digital method uses Independent Component Analysis (ICA) to generate melanin-based images representing skin areas due to melanin followed by Region Growing process to segment vitiligo lesion from normal skin.

RESULTS

Based on 41 digital images of vitiligo lesions taken from 18 patients, the proposed method achieved sensitivities of 0.9105 ± 0.0161, specificities of 0.9973 ± 0.0009 and accuracies of 0.9901 ± 0.0028 at 95% confidence level.

CONCLUSION

With the proposed method, physicians are able to assess vitiligo treatment efficacies objectively.

Intrinsic melanin and hemoglobin colour components for skin lesion malignancy detection

In this paper we propose a new log-chromaticity 2-D colour space, an extension of previous approaches, which succeeds in removing confounding factors from dermoscopic images: (i) the effects of the particular camera characteristics for the camera system used in forming RGB images; (ii) the colour of the light used in the dermoscope; (iii) shading induced by imaging non-flat skin surfaces; (iv) and light intensity, removing the effect of light-intensity falloff toward the edges of the dermoscopic image. In the context of a blind source separation of the underlying colour, we arrive at intrinsic melanin and hemoglobin images, whose properties are then used in supervised learning to achieve excellent malignant vs. benign skin lesion classification. In addition, we propose using the geometric-mean of colour for skin lesion segmentation based on simple grey-level thresholding, with results outperforming the state of the art.

Noninvasive imaging of human skin hemodynamics using a digital red-green-blue camera

In order to visualize human skin hemodynamics, we investigated a method that is specifically developed for the visualization of concentrations of oxygenated blood, deoxygenated blood, and melanin in skin tissue from digital RGB color images. Images of total blood concentration and oxygen saturation can also be reconstructed from the results of oxygenated and deoxygenated blood. Experiments using tissue-like agar gel phantoms demonstrated the ability of the developed method to quantitatively visualize the transition from an oxygenated blood to a deoxygenated blood in dermis. In vivo imaging of the chromophore concentrations and tissue oxygen saturation in the skin of the human hand are performed for 14 subjects during upper limb occlusion at 50 and 250 mm Hg. The response of the total blood concentration in the skin acquired by this method and forearm volume changes obtained from the conventional strain-gauge plethysmograph were comparable during the upper arm occlusion at pressures of both 50 and 250 mm Hg. The results presented in the present paper indicate the possibility of visualizing the hemodynamics of subsurface skin tissue.

Hyperspectral optical imaging of human iris in vivo: characteristics of reflectance spectra

We report a hyperspectral imaging system to measure the reflectance spectra of real human irises with high spatial resolution. A set of ocular prosthesis was used as the control condition. Reflectance data were decorrelated by the principal-component analysis. The main conclusion is that spectral complexity of the human iris is considerable: between 9 and 11 principal components are necessary to account for 99% of the cumulative variance in human irises. Correcting image misalignments associated with spontaneous ocular movements did not influence this result. The data also suggests a correlation between the first principal component and different levels of melanin present in the irises. It was also found that although the spectral characteristics of the first five principal components were not affected by the radial and angular position of the selected iridal areas, they affect the higher-order ones, suggesting a possible influence of the iris texture. The results show that hyperspectral imaging in the iris, together with adequate spectroscopic analyses provide more information than conventional colorimetric methods, making hyperspectral imaging suitable for the characterization of melanin and the noninvasive diagnosis of ocular diseases and iris color.

Retrieving skin properties from in vivo spectral reflectance measurements

A previously developed inverse method was applied to in vivo normal-hemispherical spectral reflectance measurements taken on the inner and outer forearm as well as the forehead of healthy white Caucasian and black African subjects. The inverse method was used to determine the thickness and melanin concentration in the epidermis, dermal blood volume fraction and oxygen saturation, and skin's spectral scattering coefficient. It was established that changes in melanin concentration due to racial difference and tanning, and differences in epidermal thickness and blood volume with anatomical location were detectable. The retrieved values were also consistent with independent measurements reported in the literature. The same method could be used for optical diagnosis of pathologies affecting the structure and pigmentation of human skin.

Quantitative principal component model for skin chromophore mapping using multi-spectral images and spatial priors

We describe a novel reconstruction algorithm based on Principal Component Analysis (PCA) applied to multi-spectral imaging data. Using numerical phantoms, based on a two layered skin model developed previously, we found analytical expressions, which convert qualitative PCA results into quantitative blood volume and oxygenation values, assuming the epidermal thickness to be known. We also evaluate the limits of accuracy of this method when the value of the epidermal thickness is not known. We show that blood volume can reliably be extracted (less than 6% error) even if the assumed thickness deviates 0.04mm from the actual value, whereas the error in blood oxygenation can be as large as 25% for the same deviation in thickness. This PCA based reconstruction was found to extract blood volume and blood oxygenation with less than 8% error, if the underlying structure is known. We then apply the method to in vivo multi-spectral images from a healthy volunteer's lower forearm, complemented by images of the same area using Optical Coherence Tomography (OCT) for measuring the epidermal thickness. Reconstruction of the imaging results using a two layered analytical skin model was compared to PCA based reconstruction results. A point wise correlation was found, showing the proof of principle of using PCA based reconstruction for blood volume and oxygenation extraction.

Principal component model of multispectral data for near real-time skin chromophore mapping

Multispectral images of skin contain information on the spatial distribution of biological chromophores, such as blood and melanin. From this, parameters such as blood volume and blood oxygenation can be retrieved using reconstruction algorithms. Most such approaches use some form of pixelwise or volumetric reconstruction code. We explore the use of principal component analysis (PCA) of multispectral images to access blood volume and blood oxygenation in near real time. We present data from healthy volunteers under arterial occlusion of the forearm, experiencing ischemia and reactive hyperemia. Using a two-layered analytical skin model, we show reconstruction results of blood volume and oxygenation and compare it to the results obtained from our new spectral analysis based on PCA. We demonstrate that PCA applied to multispectral images gives near equivalent results for skin chromophore mapping and quantification with the advantage of being three orders of magnitude faster than the reconstruction algorithm.

Rapid and accurate estimation of blood saturation, melanin content, and epidermis thickness from spectral diffuse reflectance

We present a method to determine chromophore concentrations, blood saturation, and epidermal thickness of human skin from diffuse reflectance spectra. Human skin was approximated as a plane-parallel slab of variable thickness supported by a semi-infinite layer corresponding to the epidermis and dermis, respectively. The absorption coefficient was modeled as a function of melanin content for the epidermis and blood content and oxygen saturation for the dermis. The scattering coefficient and refractive index of each layer were found in the literature. Diffuse reflectance spectra between 490 and 620 nm were generated using Monte Carlo simulations for a wide range of melanosome volume fraction, epidermal thickness, blood volume, and oxygen saturation. Then, an inverse method was developed to retrieve these physiologically meaningful parameters from the simulated diffuse reflectance spectra of skin. A previously developed accurate and efficient semiempirical model for diffuse reflectance of two layered media was used instead of time-consuming Monte Carlo simulations. All parameters could be estimated with relative root-mean-squared error of less than 5% for (i) melanosome volume fraction ranging from 1% to 8%, (ii) epidermal thickness from 20 to 150 mum, (iii) oxygen saturation from 25% to 100%, (iv) blood volume from 1.2% to 10%, and (v) tissue scattering coefficient typical of human skin in the visible part of the spectrum. A similar approach could be extended to other two-layer absorbing and scattering systems.

Three-dimensional object-distortion-tolerant recognition for integral imaging using independent component analysis

Independent component analysis (ICA) aims at extracting unknown components from multivariate data assuming that the underlying components are mutually independent. This technique has been successfully applied to the recognition and classification of objects. We present a method that combines the benefits of ICA and the ability of the integral imaging technique to obtain 3D information for the recognition of 3D objects with different orientations. Our recognition is also possible when the 3D objects are partially occluded by intermediate objects.

Image-based control of skin translucency

We propose a method for skin translucency control of facial images. This is one of the important tasks in the reproduction of posters, TV commercials, movies, and so on. As the first step of processing, we extract the component maps of melanin, hemoglobin, and shading from skin color images by using our conventional method. The extracted shading component is controlled to change the translucency of the skin by simple kernel operations for the component. The efficiency for the change of translucency is confirmed by using the images of numerical and optical skin phantoms. The method is also applied into the real skin color image with the consideration of each component, and realistic change of skin translucency was observed from the resultant images synthesized by the proposed method.

Determination of skin repigmentation progression

In this paper, we describe an image processing scheme to analyze and determine areas of skin that have undergone repigmentation in particular, during the treatment of vitiligo. In vitiligo cases, areas of skin become pale or white due to the lack of skin pigment called melanin. Vitiligo treatment causes skin repigmentation resulting in a normal skin color. However, it is difficult to determine and quantify the amount of repigmentation visually during treatment because the repigmentation progress is slow and moreover changes in skin color can only be discerned over a longer time frame typically 6 months. Here, we develop a digital image analysis scheme that can identify and determine vitiligo skin areas and repigmentation progression on a shorter time period. The technique is based on principal component analysis and independent component analysis which converts the RGB skin image into a skin image that represent skin areas due to melanin and haemoglobin only, followed by segmentation process. Vitiligo skin lesions are identified as skin areas that lack melanin (non-melanin areas). In the initial studies of 4 patients, the method has been able to quantify repigmentation in vitiligo lesion. Hence it is now possible to determine repigmentation progression objectively and treatment efficacy on a shorter time cycle.

Visualizing depth and thickness of a local blood region in skin tissue using diffuse reflectance images

A method is proposed for visualizing the depth and thickness distribution of a local blood region in skin tissue using diffuse reflectance images at three isosbestic wavelengths of hemoglobin: 420, 585, and 800 nm. Monte Carlo simulation of light transport specifies a relation among optical densities, depth, and thickness of the region under given concentrations of melanin in epidermis and blood in dermis. Experiments with tissue-like agar gel phantoms indicate that a simple circular blood region embedded in scattering media can be visualized with errors of 6% for the depth and 22% for the thickness to the given values. In-vivo measurements on human veins demonstrate that results from the proposed method agree within errors of 30 and 19% for the depth and thickness, respectively, with values obtained from the same veins by the conventional ultrasound technique. Numerical investigation with the Monte Carlo simulation of light transport in the skin tissue is also performed to discuss effects of deviation in scattering coefficients of skin tissue and absorption coefficients of the local blood region from the typical values of the results. The depth of the local blood region is over- or underestimated as the scattering coefficients of epidermis and dermis decrease or increase, respectively, while the thickness of the region agrees well with the given values below 1.2 mm. Decreases or increases of hematocrit value give over- or underestimation of the thickness, but they have almost no influence on the depth.

Inspection of skin hemodynamics with hyperspectral camera

Using hyperspectral imaging techniques, which enable us to simultaneously obtain spatial and wavelength information, we have improved upon the newly developed hyperspectral camera to develop a method of observing changes in skin melanin levels and hemodynamics over time. As an evaluation of this method, we measured the skin diffuse reflectance spectrum of the human middle finger in an experiment of blood flow blockage in the brachial region. The changes in skin hemodynamics observed through this method match the behavior expected based on clinical knowledge, and also show an extremely high correlation with results obtained by the Erythema Index, which is used to make similar calculations from a limited number of wavelengths.

Identification of pure component spectra by independent component analysis in glucose prediction based on mid-infrared spectroscopy

We present a method for glucose prediction from mid-IR spectra by independent component analysis (ICA). This method is able to identify pure, or individual, absorption spectra of constituent components from the mixture spectra without a priori knowledge of the mixture. This method was tested with a two-component system consisting of an aqueous solution of both glucose and sucrose, which exhibit distinct but closely overlapped spectra. ICA combined with principal component analysis was able to identify a spectrum for each component, the correct number of components, and the concentrations of the components in the mixture. This method does not need a calibration process and is advantageous in noninvasive glucose monitoring since expensive and time-consuming clinical tests for data calibration are not required.

A subspace matching color filter design methodology for a multispectral imaging system

In this paper, we present a methodology to design filters for an imaging system to improve the accuracy of the spectral measurements for families of reflective surfaces. We derive the necessary and sufficient conditions that the sensor space of the system must obey in order to measure the spectral reflectance of the surfaces accurately. Through simulations, we show how these conditions can be applied to design filters using a set of sample spectral data acquired from extracted teeth. For this set of data, we also compare our results to those of Wolski's method, a conventional filter design method which produces filters that recover tristimulus values of surfaces accurately under several illuminants. We show that our method produces filters that capture the spectral reflectance better given the same number of measurements. The errors in predicting the color of the sample data are much lower under every test illuminant when the filters designed with our method are used.

Image-based control of skin melanin texture

We introduce a useful tool for controlling the skin melanin texture of facial photographs. Controlling the skin melanin texture is an important task in the reproduction of posters, TV commercials, movies, and so on. We used component maps of melanin, which were obtained by a previous method [J. Opt. Soc. Am. A 16, 2169 (1999)] as the first processing step. We propose to control the melanin texture continuously and physiologically, based on the analysis of 123 skin textures in our database. The physiological validity for the change of the melanin texture is confirmed by comparing the synthesized image with an ultraviolet image, which can be used to predict the change of melanin texture due to aging. The control processes are implemented on programmable graphics hardware, and real-time processing is achieved for a facial videostream.

Interpretation of principal components of the reflectance spectra obtained from multispectral images of exposed pig brain

The spatial variation in reflectance such as the blood-vessel pattern can be observed in the image of cerebral cortex. This spatial variation is mainly caused by the difference in concentrations of oxy- and deoxyhemoglobin in the tissue. We analyze the reflectance spectra obtained from multispectral images of pig cortex by principal component analysis to extract information that relates to physiological parameters such as the concentrations of oxy- and deoxyhemoglobin and physical parameters such as mean optical path length. The light propagation in a model of exposed pig cortex is predicted by Monte Carlo simulation to estimate the interpretation of physiological and physical meanings of the principal components. The spatial variance of reflectance spectra of the pig cortex can be approximately described by the first principal component. The first principal component reflects the spectrum of hemoglobin in the cortical tissue multiplied by the mean optical path length. These results imply that the wavelength dependence of mean optical path length can be experimentally estimated from the first principal component of the reflectance spectra obtained from multispectral image of cortical tissue.